SIMO based DC-DC converters for thermoelectric energy harvesting

Abstract

A single-inductor multiple-output (SIMO) power converter converts an input voltage into an output voltage and a biasing voltage. The SIMO power converter comprises an inductor and a primary power switch, and a control circuit. The inductor is configured for storing energy from the input voltage. The primary power switch has a control node and is connected between the inductor and the output voltage which powers an output load. The control circuit controls the primary power switch comprising an auxiliary power switch and a driver. The auxiliary power switch is connected between the inductor and the biasing voltage. The driver, powered by the biasing voltage, drives the control node. The biasing voltage determines a signal level at the control node. The primary power switch and the auxiliary power switch are controlled to distribute the energy stored in the inductor to the output voltage and the biasing voltage.

Description

CROSS-REFERENCE TO RELATED APPLICATION

This application claims priority to and the benefit of Taiwan Application Series Number 104102059 filed on Jan. 22, 2015, which is incorporated by reference in its entirety.

BACKGROUND

The present disclosure relates generally to single-inductor multiple-output (SIMO) based DC-DC converters, more especially to SIMO based converters with adaptive gate biasing (AGB) technique for thermoelectric energy harvesting.

Energy harvested from the environment can be used to develop battery-free electronics systems or prolong battery life. Among the different green energy sources from ambient environment such as light, motion, and heat, thermal power from human body is an efficient and reliable energy source for wearable applications. However, the output voltage of thermoelectric generators (TEGs) is typically less than 100 mV for a thermal difference of 2K depending on the temperature dependent output characteristics range of 10 mV/K to 50 mV/K. Moreover, considering the limited power budget of TEGs, the load system typically requires digital circuits operating in the near-threshold region to reduce power dissipation. Therefore, a power converter that can convert the harvested energy to a near-threshold output is required to realize an energy efficient system. However, designing a high efficiency low-V_IN low-V_OUT converter is challenging owing to the significant conduction losses (P_CONDUCTION) in power transistors.

Numerous thermoelectric energy harvesting power converters have been proposed in the art for low V_IN and low power operation. However, the output stages of these power converters are greater than 0.9V. A two-stage topology with a cascaded auxiliary boost converter and a DC-DC buck converter has been proposed, as shown in FIG. 1. An auxiliary boost converter with an additional off-chip inductor L_AUX acts as a buffer, and the DC-DC buck converter maintains the V_OUT value at 1.8V. With this two-stage topology, the low V_IN and low V_OUT specifications can be realized because V_OUT can be regulated to any desired level. However, this structure in FIG. 1 has low efficiency because of the two-stage conversion and high conversion ratio in the auxiliary boost converter.

BRIEF DESCRIPTION OF THE DRAWINGS

Non-limiting and non-exhaustive embodiments of the present invention are described with reference to the following drawings. In the drawings, like reference numerals refer to like parts throughout the various figures unless otherwise specified. These drawings are not necessarily drawn to scale. Likewise, the relative sizes of elements illustrated by the drawings may differ from the relative sizes depicted.

The invention can be more fully understood by the subsequent detailed description and examples with references made to the accompanying drawings, wherein:

FIG. 1 shows a thermoelectric energy harvesting power converter in the art;

FIG. 2 shows a single-stage power converter according to embodiments of the invention;

FIG. 3 shows a conceptual block diagram of the AGB technique;

FIG. 4A shows the circuit schematic of the primary boost converter 20 in FIG. 3;

FIG. 4B demonstrates the conversion efficiencies of converters with AGE, technique and without AGB technique respectively;

FIG. 5 shows the primary boost converter and the biasing voltage generator 14 in FIG. 3;

FIG. 6 demonstrates the PWM generator in FIG. 3, including a latch comparator and a 7-bit DPWM (digital PWM) generator;

FIG. 7 details the P_OUT detector and the level-shift module in FIG. 3;

FIGS. 8A and 8B demonstrates the latch comparators PD1 and PD2, respectively;

FIG. 9 shows four different operation statuses PathA, PathB, PathC, and PathD for the SIMO of FIG. 5;

FIGS. 10A, 10B, 10C and 10D show the inductor current waveforms when the SIMO of FIG. 5 operates at the four statuses, respectively; and

FIG. 11 shows the energy delivery mechanism of near-V_TH ERC and skipping modulation in response to skip signals S_KIP1 and S_KIP2.

DETAILED DESCRIPTION

An embodiment of the invention introduces a 100 mV V_IN, 500 mV V_OUT high-efficiency SIMO-based boost converter for harvesting thermoelectric energy generated by thermoelectric generators. FIG. 2 shows a single-stage power converter according to embodiments of the invention, which employs adaptive gate biasing (AGB) technique and near-threshold voltage (near-V_TH) energy redistribution control (ERC). To improve the conversion efficiency at low V_OUT, two key techniques are used. (1) An AGB technique that provides dual load-dependent voltages (gate biasing voltages V_AGB1 and V_AGB2) for driving the power transistors of the primary boost converter. This technique improves the conversion efficiency by reducing P_CONDUCTION and switching loss (P_SWITCHING) at different load conditions and providing a self-calibration mechanism against V_TH variation. (2) A near-V_TH ERC mechanism that is powered by V_OUT and manages the power delivery strategy of output voltage V_OUT, V_AGB1, and V_AGB2. One embodiment of the invention achieves a maximum efficiency of 83.4% at an output power (P_OUT) of 250 uW, and power efficiency greater than 80% over the output power range of 150 uW to 450 uW. In addition, the core controller is implemented using 7-bit delay-line based digital pulse width modulation (DPWM), which not only reduces the quiescent power to 0.48 uW but also ensures reliable digital operation with the near-V_TH power supply form output voltage V_OUT.

FIG. 3 shows a conceptual block diagram of the AGB technique. PWM generator 18 in the AGB circuit 10 provides digital PWM signal CK_PWM based on a clock signal CK, output voltage V_OUT and reference voltage V_REF. The P_OUT detector 12, powered by output voltage V_OUT, senses the voltage difference between V_X and V_SS when M_N1 is ON, and the voltage difference between V_X and V_OUT when M_N2 is ON. As indicated in FIG. 3, V_SS is the voltage at a ground line and V_X is at the joint node between M_N2 and M_N1. Outputs V_GP1 and V_GP2 of the P_OUT detector 12 control the biasing voltage generator 14 to generate gate biasing voltages V_AGB1 and V_AGB2 which perform as power sources for level-shift and driver module 16. Based on the PWM signal CK_PWM and outputs of the P_OUT detector 12, level-shifter and driver module 16 provides gate control signals V_GN1 and V_GN2, whose signal levels are determined by gate biasing voltages V_AGB1 and V_AGB2 respectively. Therefore, this AGB technique can reduce P_CONDUCTION and P_SWITCHING by providing appropriate gate overdrive voltages. It also precludes output voltage V_OUT from a tradeoff between output load demand and the performance of the power converter and compensates for threshold voltage V_TN shift due to process or temperature variation. The operation principle and implementation of the AGB technique are described in detail below.

FIG. 4A shows the circuit schematic of the primary boost converter 20 in FIG. 3. This primary boost converter 20 includes an inductor L_BOOST N-type power transistors M_N2 and M_N2, and the output capacitor C_OUT. This primary boost converter 20 stores energy when the power transistor M_N2 turns ON (M_N2 turns OFF) to operate under status PhaseA. The stored energy is then delivered to output voltage V_OUT when the power transistor M_N2 turns ON (M_N2 turns OFF) to operate at status PhaseB. In one embodiment, because output voltage V_OUT is targeted to be 0.5V, in the near-V_TH region, an N-type power transistor M_N2 is employed as the high-side power transistor. Compared with a P-type transistor, an N-type power transistor has higher mobility and superior area efficiency. In addition, only a positive voltage is required for driving an N-type power transistor whereas a voltage ranging from positive to negative potential is essential for a P-type transistor to effectively implement the AGB algorithm. Therefore, in one embodiment, no additional negative voltage converter is required and implemented, and the complexity of the circuit design is reduced.

To simplify the analysis, it is assumed that the primary boost converter 20 operates under continuous conduction mode (CCM), and the inductor current ripple is negligible. In a steady state, P_CONDUCTION of power transistors M_N2 and M_N2 can be expressed as

$\begin{array}{cc}\begin{array}{c}{P}_{\mathrm{CONDUCTION}}=P_{\mathrm{MN}1}+P_{\mathrm{MN}2}\\ ={\int }_0^{\mathrm{DT}}I_L^2R_{\mathrm{DS}1,\mathrm{ON}}t+{\int }_{\mathrm{DT}}^TI_L^2R_{\mathrm{DS}2,\mathrm{ON}}t,\end{array}& \left(1\right)\end{array}$

where I_L is the average inductor current flowing through the inductor L_BOOST, R_DS1,ON and P_DS2,ON are the on-resistances of power transistors M_N1 and M_N2r respectively. Further, T is the clock period, and D is the duty cycle of the signal provided to the power transistor M_N1 in a steady state. For a boost converter, the relation between I_L and the output load current I_LOAD can be expressed as

$\begin{array}{cc}{I}_L=\frac{I_{\mathrm{LOAD}}}{1-D}.& \left(2\right)\end{array}$

Therefore, the correlation between P_CONDUCTION and I_LOAD can be derived as

$\begin{array}{cc}\begin{array}{c}{P}_{\mathrm{CONDUCTION}}={\int }_0^{\mathrm{DT}}{\left(\frac{I_{\mathrm{LOAD}}}{1-D}\right)}^2R_{\mathrm{DS}1,\mathrm{ON}}t+\\ {\int }_{\mathrm{DT}}^T{\left(\frac{I_{\mathrm{LOAD}}}{1-D}\right)}^2R_{\mathrm{DS}2,\mathrm{ON}}t\\ =I_{\mathrm{LOAD}}^2\left(R_{\mathrm{DS}1,\mathrm{ON}}\frac{\mathrm{DT}}{{\left(1-D\right)}^2}+R_{\mathrm{DS}2,\mathrm{ON}}\frac{T}{1-D}\right).\end{array}& \left(3\right)\end{array}$

Hence, if the duty cycle, operation frequency, fabrication process and transistor size of the power transistors are known, P_CONDUCTION is in proportion to the square of I_LOAD and increases dramatically as I_LOAD increases, as given by the equation (3).

By detecting the voltage drop across the power transistors and adjusting the corresponding gate biasing voltages, the proposed AGB technique maintains the turn-on voltages V_DS1,ON and V_DS2,ON to be about constant.

$\begin{array}{cc}{V}_{\mathrm{DS}1,\mathrm{ON}}=I_LR_{\mathrm{DS}1,\mathrm{ON}}=\frac{I_{\mathrm{LOAD}}}{1-D}R_{\mathrm{DS}1,\mathrm{ON}}.& \left(4\right)\\ V_{\mathrm{DS}2,\mathrm{ON}}=I_LR_{\mathrm{DS}2,\mathrm{ON}}=\frac{I_{\mathrm{LOAD}}}{1-D}R_{\mathrm{DS}1,\mathrm{ON}}.& \left(5\right)\end{array}$

Therefore, the total conduction loss can be derived from equations (3), (4), and (5) as

$\begin{array}{cc}{P}_{\mathrm{CONDUCTION}}=I_{\mathrm{LOAD}}\left(V_{\mathrm{DS}1,\mathrm{ON}}\frac{\mathrm{DT}}{1-D}+V_{\mathrm{DS}2,\mathrm{ON}}·T\right).& \left(6\right)\end{array}$

If turn-on voltages V_DS1,ON and V_DS2,ON are constant, P_CONDUCTION is proportional only to I_LOAD in equation (6) instead of the square function in the conventional equation (3). Therefore, under heavy loads, the conversion efficiency can be improved by employing the AGB technique.

On the other hand, under light-load conditions, P_CONDUCTION decreases due to the reduced I_L flowing through power transistors, and P_SWITCHING dominates the overall efficiency. The AGB technique detects the decreasing turn-on voltages V_DS1,ON and V_DS2,ON, and the potentials of V_AGB1 and V_AGB2 are simultaneously reduced to maintain the turn-on voltages V_DS1,ON and V_DS2,ON across the power transistors M_N1 and M_N2. Therefore, lower values of gate biasing voltages V_AGB1 and V_AGB2 decrease P_SWITCHING as given by

P_SWITCHING=f_CKC_GATE1V_AGB1²+f_CKC_GATE2V_AGB2²  (7)

where f_CK is the operation frequency (=1/T), and C_GATE1 and C_GATE2 are the gate capacitances of the power transistors M_N1 and M_N2, respectively. Because P_SWITCHING is proportional to the square of the gate biasing voltages, V_AGB1 and V_AGB2 power efficiency is predicted to be higher with the AGB technique under light load. Therefore, by applying the AGB technique, both light-load and heavy-load efficiencies can be improved, as shown in FIG. 4B. The AGB technique automatically adjusts gate biasing voltages V_AGB1 and V_AGB2 according to I_LOAD, so as to manipulate the on-resistances R_DS1,ON and R_DS2,ON, thus resulting in significant suppression of losses under different load conditions.

AGB circuit 10 in FIG. 3 implements the AGB technique, providing a feedback mechanism to regulate or maintain both turn-on voltages V_DS1,ON and V_DS2,ON to be about constant, independent from the load condition.

FIG. 5 shows the primary boost converter 20 and the biasing voltage generator 14 in FIG. 3, both composing a SIMO to simultaneously generate output voltage V_OUT, and gate biasing voltages V_AGB1 and V_AGB2 Capacitors C₁ and C₂ are used to stabilize gate biasing voltages V_AGB1 and V_AGB2 Inductor L_BOOST N-type power transistors M_N1 and M_N2, and P-type power transistor M_P1 and M_P2 are responsible for the energy stored and released in different time slots within a period T. Energy is delivered from input voltage V_IN via being stored in inductor L_BOOST, and then distributed to the three output terminals through time-multiplexing control accompanied with skipping modulation, which precludes output voltage V_OUT from the crosstalk between gate biasing voltages V_AGB1 and V_AGB2 and will be discussed later.

FIG. 6 demonstrates the PWM generator 18 in FIG. 3, including a latch comparator 22 and a 7-bit DPWM (digital PWM) generator 24. The latch comparator 22 compares output voltage V_OUT with a reference voltage V_REF, which is 0.5V for this embodiment, and controls the duty cycle of DPWM signal CK_PWM via the high-resolution DPWM unit 24. For instance, if output voltage V_OUT is below reference voltage V_REF the duty cycle of DPWM signal CK_PWM increases stepwise by a fixed amount. Accordingly, in a steady state, output voltage V_OUT could be regulated at the reference voltage V_REF To reduce power consumption by DPWM unit 24 and the driver circuit, a clock frequency of 100 kHz is selected in this embodiment, as demonstrated by the reference clock signal CK in FIG. 6.

FIG. 7 details the P_OUT detector 12 and level-shift and driver module 16 of FIG. 3. The P_OUT detector 12 has latch comparators PD1 and PD2, and ERC unit 26, while the level-shift and driver module 16 has level shifters 28 and 30, and drivers 32 and 34.

The latch comparator PD1 detects the voltage difference between V_X and V_SS during the ON time of the power transistor M_N1. In other words, it detects the turn-on voltage V_DS1,ON, with which a predetermined value V_OFFSET1 is compared to generate skip signal S_KIP1 Similarly, the latch comparator PD2 detects the voltage difference between V_X and output voltage V_OUT, which is the turn-on voltage V_DS2,ON Turn-on voltage V_DS2,ON is compared with a predetermined value V_OFFSET2 to generate skip signal S_KIPS. When S_KIP1 is “1” in logic, it means turn-on voltage V_DS1,ON is below or equal to the predetermined value V_OFFSET1. When S_KIP1 is “0” in logic, turn-on voltage V_DS1,ON exceeds the predetermined value V_OFFSET1. Skip signal S_KIP2 in logic “1” means the turn-on voltage V_DS2,ON is below or equal to the predetermined value V_OFFSET2, while that in logic “0” means the turn-on voltage V_DS2,ON exceeds the predetermined value V_OFFSET2. The latch comparators PD1 and PD2 are illustrated in FIGS. 8A and 8B, respectively. The predetermined value V_OFFSET1 is implemented using the intentionally mismatched PMOSFET input pair in FIG. 8A, and the predetermined value V_OFFSET2 is implemented using the intentionally mismatched NMOSFET input pair in FIG. 8B. In FIG. 7, ERC unit 26 decides whether to pull low gate control signals V_GP1 and V_GP2 at corresponding time slots, depending on skip signals S_KIP1 and S_KIP2.

FIG. 9 shows four different operation statuses PathA, PathB, PathC, and PathD for the SIMO of FIG. 5. FIGS. 10A, 10B, 10C and 10D show the inductor current waveforms when the SIMO of FIG. 5 operates at the four statuses, respectively. Under operation status PhaseA, power transistor M_N1 turns ON and the energy is stored from input voltage V_IN in inductor L_BOOST, and the duty cycle CK_PWM for this time slot is determined by the 7-bit DPWM unit 24. Part of the energy accumulated in status PhaseA is then distributed to gate biasing voltage V_AGB2 under the status PhaseD, via the turn-on of power transistor M_P2, where gate biasing voltage V_AGB2 is the highest voltage in this system for this embodiment. Sequentially and similarly, the energy accumulated in status PhaseA is partially provided for gate biasing voltage V_AGB1 under the status PhaseC, via the turn-on of power transistor M_P1. Finally, output voltage V_OUT receives the rest of the accumulated energy under status PhaseB to fulfill the output load. Even though the status sequence shown in FIGS. 10A to 10D is (PhaseA, PhaseD, PhaseC, PhaseB), this invention is not limited to, and a different status sequence is possibly employed in another embodiment. Statuses PhaseD and PhaseC are necessary for generating adaptive gate biasing voltages V_AGB1 and V_AGB2; however, they inevitably cause discontinuous conduction and disturbance for primary output voltage V_OUT. To overcome this problem, the pulse width of the driving signals provided for power transistors M_P1 and M_P2 is limited to 100 ns to restrict the amount of power transferred to gate biasing voltages V_AGB1 and V_AGB2. In other words, if any of power transistors M_P1 and M_P2 turns ON, the ON time will be always 100 ns. Additionally, status PhaseD is activated only when the turn-on voltage V_DS2,ON exceeds the predetermined value V_OFFSET2 while status PhaseC is activated only when the turn-on voltages V_DS1,ON exceeds the predetermined value V_OFFSET1.

The power consumed by gate biasing voltages V_AGB1 and V_AGB2 is typically less than one percent of that of the primary output voltage V_OUT. Therefore, the ERC unit 26 in FIG. 7 implements a skipping modulation mode to avoid large disparity in load conditions among V_AGB1, V_AGB2 and V_OUT. FIG. 11 shows the energy delivery mechanism of near-V_TH ERC and skipping modulation in response to skip signals S_KIP1 and S_KIP2 Generally, three output terminals receive energy to maintain their voltage levels, as shown in mode IV when skip signals S_KIP1 and S_KIP2 are both “0” in logic. In the mode III when skip signals S_KIP1 and S_KIP2 are “0” and “1” respectively, the status PhaseD is skipped and gate biasing voltage V_AGB2 does not receive energy. Similarly, In the mode II when skip signals S_KIP1 and S_KIP2 are “1” and “0” respectively, the status PhaseC is skipped and gate biasing voltage V_AGB1 does not receive energy. When both the gate biasing voltages V_AGB1 and V_AGB2 are high enough to keep the turn-on voltages V_DS1,ON and V_DS2,ON below or equal to the predetermined values V_OFFSET1 and V_OFFSET2 respectively, P-type power transistors M_P1 and M_P2 are constantly kept OFF, and both the statuses PhaseC and PhaseD are skipped, as shown in the mode I. If the status PhaseC (PhaseD) is skipped, the gate biasing voltages V_AGB1 (V_AGB2) ramps down due to P_SWITCHING, and the turn-on voltage V_DS1,ON (V_DS2,ON) in a subsequent switch cycle increases as a result. Therefore, by appropriate energy delivery management, the proposed near-V_T ERC and skipping modulation technique ensure that each of three voltages is sufficiently isolated and independent of the others to stabilize the entire power converter.

It is shown by the mode IV in FIG. 11 that the 100 nS after status PhaseA is for status PhaseD, the 100 nS after which is for status PhaseC, which is followed by status PhaseB.

While the invention has been described by way of example and in terms of preferred embodiment, it is to be understood that the invention is not limited thereto. To the contrary, it is intended to cover various modifications and similar arrangements (as would be apparent to those skilled in the art). Therefore, the scope of the appended claims should be accorded the broadest interpretation so as to encompass all such modifications and similar arrangements.

Claims

1. A single-inductor multiple-output (SIMO) power converter for converting an input voltage, comprising:

an inductor, configured for storing energy from the input voltage;

a primary power switch with a control node, connected between the inductor and an output voltage, wherein the output voltage is for powering an output load; and

a control circuit for controlling the primary power switch, the control circuit comprising: an auxiliary power switch, connected between the inductor and a biasing voltage; and a driver powered by the biasing voltage, for driving the control node, wherein the biasing voltage determines a signal level at the control node; wherein the primary power switch and the auxiliary power switch are controlled to store the energy in the inductor and distribute to the output voltage and the biasing voltage.

2. The SIMO power converter of claim 1, wherein the energy is distributed in response to a turn-on voltage of the primary power switch to control the signal level.

3. The SIMO power converter of claim 2, wherein the control circuit provides a negative feedback mechanism for regulating the turn-on voltage to be a predetermined value.

4. The SIMO power converter of claim 3, wherein when the turn-on voltage exceeds the predetermined value the auxiliary power switch turns ON at a predetermined time slot during a switching period.

5. The SIMO power converter of claim 3, wherein when the turn-on voltage is below or equal to the predetermined value the control circuit stops distributing the energy to the biasing voltage.

6. The SIMO power converter of claim 1, comprising a boost converter powered by the input power, wherein both the output voltage and the biasing voltage exceed the input voltage.

7. The SIMO power converter of claim 6, wherein the primary power switch is an N-type power transistor, and the auxiliary power switch is a P-type power transistor.

8. The SIMO power converter of claim 6, wherein the auxiliary power switch is a second auxiliary power switch, the biasing voltage is a second biasing voltage, the driver is a second driver, and the primary power switch is a high-side power switch, the SIMO power converter further comprising:

a low-side power switch connected between the inductor and a ground line for making the inductor store the energy when the low-side power switch turns on; and

a first auxiliary power switch connected between the inductor and a first biasing voltage;

wherein the high-side power switch, the first auxiliary power switch and the second auxiliary power switch are controlled to distribute the energy stored in the inductor to the output voltage, the first biasing voltage, and the second biasing voltage;

the control circuit further comprises a first driver powered by the first biasing voltage to control the low-side power switch; and

the first biasing voltage determines a signal level at a control node of the low-side power switch.

9. The SIMO power converter of claim 8, wherein the energy is distributed in response to a turn-on voltage of the low-side power switch, so as to regulate the turn-on voltage of the low-side power switch to be a predetermined value.

10. The SIMO power converter of claim 8, further comprising a PWM generator determining a duty cycle of the low-side power switch in response to the output voltage and a reference voltage.

11. The SIMO power converter of claim 10, wherein the PWM generator comprises:

a latch comparator for comparing the output voltage with the reference voltage; and

a digital PWM generator for adjusting the duty cycle according to an output from the latch comparator.

12. The SIMO power converter of claim 8, wherein the control circuit comprises first and second power detectors, the high-side and the low-side power switches are connected via a joint node, the first power detector detects a voltage difference between the ground line and the joint node to control the first driver, and the second power detector detects a voltage difference between the output voltage and the joint node to control the second driver.

13. A method suitable for a single-inductor multiple-output (SIMO) power converter, wherein the SIMO power converter comprises a primary power switch connected between an inductor and an output voltage, and an auxiliary power switch between the inductor and a biasing voltage, the method comprising:

powering an output load by using the output voltage;

regulating the output voltage at a reference voltage;

sensing a turn-on voltage across the primary power switch; and

providing a pulse signal to control the auxiliary power switch so as to maintain the turn-on voltage to be about a predetermined value;

wherein the biasing voltage determines a signal level at a control node of the primary power switch.

14. The method of claim 13, further comprising

comparing the turn-on voltage with the predetermined value by using a latch comparator with a mismatched input pair.

15. The method of claim 13, wherein when the turn-on voltage is below or equal to the predetermined value the pulse signal is not provided and the auxiliary power switch is constantly turned off.

16. The method of claim 15, wherein, when the turn-on voltage exceeds the predetermined value, the pulse signal is provided during a predetermined time slot so as to turn ON the auxiliary power switch during the predetermined time slot.

17. The method of claim 13, wherein the step of regulating the output voltage comprises:

providing a PWM signal based on a difference between the output voltage and the reference voltage.

18. The method of claim 17, wherein the PWM signal is provided by a digital PWM generator according to an output from a latch comparator which compares the output voltage with the reference voltage.

19. The method of claim 13, wherein the SIMO power converter is a boost converter, the primary power switch is a high-side power switch, the SIMO power converter further comprises a low-side power switch between a ground line and the inductor, and a second auxiliary power switch between the inductor and a second biasing voltage, and the method further comprises:

sensing a second turn-on voltage across the low-side power switch; and

providing a second pulse signal to control the second auxiliary power switch so as to maintain the second turn-on voltage to be about a second predetermined value;

wherein the second biasing voltage determines a signal level at a control node of the second auxiliary power switch.

20. The method of claim 19, wherein the high-side and low-side power switches are NMOSFETs, and the first and second auxiliary power switched are PMOSFETs.